Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 11709
    Prev
    Next
    INDEX
    Explanations

    references to medical conditions and treatments, particularly related to diabetes and esophageal procedures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    235
    +0.18
    1.0%
    271
    +0.16
    1.0%
    47
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.18
    0.11
    312
    +0.16
    0.11
    326
    +0.15
    0.05
    Negative Logits
     shelves
    -1.63
     spreadsheet
    -1.41
    ://
    -1.40
     probe
    -1.39
     table
    -1.38
     []{
    -1.38
     bedside
    -1.38
     dash
    -1.38
     investigators
    -1.37
    ********************************
    -1.37
    POSITIVE LOGITS
    andy
    1.77
    ustom
    1.43
    ivid
    1.42
    ép
    1.37
    mapsto
    1.37
     Ages
    1.34
    ever
    1.32
    aiser
    1.30
    ivery
    1.28
    err
    1.28
    Activations Density 3.450%

    No Known Activations